Annotation, phylogenetic analysis and functional characterization of barley [Hordeum vulgare L] WRKY transcription factors in the interaction with powdery mildew fungus Blumeria graminis

(1)

Annotation, phylogenetic analysis and functional

characterization of barley [Hordeum vulgare L.]

WRKY transcription factors in the interaction with

powdery mildew fungus Blumeria graminis

Dissertation for the Achievement of the Degree “Doktor der Agrarwissenschaften”

at the Faculty of Agricultural and Nutritional Sciences, Home Economics and Environmental Management

Justus-Liebig-Universität Gießen

Performed at

Institute of Phytopathology and Applied Zoology

Submitted by

Dilin Liu

from China

Supervised by

1. Prof. Dr. Karl-Heinz Kogel 2. Prof. Dr. Andreas Vilcinskas

(2)

Table of Abbreviations

amiR artificial microRNA AVR avirulence

Bgh Blumeria graminis f.sp. hordei

BLAST basic local alignment search tool

bp base pair

CC coiled-coil

cDNA complementary DNA

ChIP chromatin immunoprecipitation cv. cultivar

CWA cell wall appositions

Da Dalton

DNA 2’-deoxy-ribonucleic acid dsRNA double-stranded RNA

E. coli Escherichia coli

EST expressed sequence tag

et.al. et altera

ETI effector triggered immunity f.sp forma specialis

GFP green fluorescent protein GLP germin-like proteins GP Golden Promise GUS β-glucuronidase H hour

HDA histone deacetylase HR hypersensitive response HSP heat shock protein

JA jasmonate

kb kilobase (s) kDa kilodalton (s) LRR leucin-rich-repeat

(6)

Table of Abbreviations

MAMP microbe-associated molecular patterns MAPK mitogen-activated protein kinase miRNA microRNA

ML mildew resistance locus MLA mildew-resistance locus A

mRNA messenger RNA

MTI MAMP-triggered immunity NBS nucleotide-binding site NLS nuclear localisation signal OD optical density

ORF open reading frame

PAGE polyacrylamide gel electrophoresis PCD programmed cell death

PCR polymerase chain reaction PR pathogenesis-related qRT-PCR quantitative real-time PCR RNA ribonucleic acid

RNAi RNA interference

Ror required for mlo-specific resistance

RT-PCR reverse transcriptase-PCR SA salicylic acid

SAR systemic acquired resistance SDS sodium dodecyl sulfate

SDS-PAGE SDS polyacrylamide gel electrophoresis SE standard error

siRNAs small interfering RNAs smRNAs small RNAs

SPF1 sweet-potato factor 1 TBE tris-borate-EDTA TE tris-EDTA

TTG2 transparent testa glabra2 VIGS virus-induced gene silencing WMD web microRNA designer

(7)

Table of Abbreviations

Xoo Xanthomonas oryzae pv. oryzae

R gene resistance gene

(8)

Introduction

1. Introduction

To cope with variable environmental conditions, plants have evolved a great capacity to extensively reprogram their transcriptome in a highly dynamic and temporal manner through an integrated network of transcription factors. Among these transcription factors, WRKY factors are a group of regulatory proteins predominantly involved in stress responses (Pandey and Somssich, 2009). Together with other transcriptional regulators, WRKY proteins enable plants to better adapt to the changing environment and respond properly to internal and external stimuli.

1.1 Structure and evolution of WRKY transcription factors 1.1.1 Structure of WRKY transcription factors

The WRKY protein family is named after the most prominent feature of these proteins, the WRKY domain, a highly conserved motif spanning about 60 amino acids in all the family members (Eulgem et al., 2000). Within this domain, there is an almost invariable heptapeptide signature WRKYGQK at the N-terminus and a novel zinc finger-like structure at the C-terminus. The WRKYGQK is the most dominant form of the signature followed by WRKYGKK and WRKYGEK (Eulgem

et al., 2000; Xie et al., 2005b), however, there are at least 35 variants of this motif

present in plant and non-plant species (Table 1.1). WRKY genes encode transcription factors and they are targeted to the nucleus as most of them contain a basic nuclear localization signal. WRKY proteins preferably bind to the consensus sequence TTGACC/T, the so-called W-box which is usually enriched in the promoter region of WRKY target genes such as stress responsive genes. Both the WRKY and zinc-finger motif are essential for proper DNA binding capacity of the protein (Maeo et al., 2001).

Table 1. 1. List of WRKY signature variants in WRKY domains from plant and non-plant species (sequence from http://supfam.cs.bris.ac.uk/SUPERFAMILY/).

WRKY

signature Distribution of WRKY variants

Number of WRKY domains

WRKYGQK All plant species 1761

(9)

Introduction

Some dicots (i.e, Glycine max, Lotus japonicus) Green algae

(Volvox carteri, Chlamydomonas reinhardtii, Coccomyxa sp. C-169)

WRKYGKK All vascular plants 88

WKKYGQK All monocots

Some dicots (i.e, Glycine max, Vitis vinefera) 16

WQKYGQK Arabidopsis thaliana,

tomato (Solanum lycopersicum) 2

WSKYGQK Tomato (Solanum lycopersicum) 1

WSKYGQM Barley (Hordeum vulgare) 1

WTKYGQK Barley (Hordeum vulgare) 1

WNKYGQK Barley (Hordeum vulgare) 1

WKRKGQK Rice (Oryza sativa) 1

WVKYGQK Rice (Oryza sativa) 1

WRRYGLK Rice (Oryza sativa) 1

WRKYEDK Soybean (Glycine max) 1

WRKYGKR Soybean (Glycine max) 1

WRKYGSK Sorghum (Sorghum bicolor),

Medicago truncatula, Giardia lamblia 3

WEKFGEK Sorghum (Sorghum bicolor) 1

WRKYGQE Wheat (Triticum aestivum) 2

WKKYGHK Giardia lamblia 1

WRKCGLK Lotus japonicus 1

WRKYGQN Lotus japonicus, Moss (Physcomitrella patens) 3

WKKYGYK Lotus japonicus 1

WKKYGED Lotus japonicus 1

WLKYGQK Lotus japonicus 1

WKKYEEK Medicago truncatula 2

WKKYGEK Medicago truncatula,

Asteraceae (Helianthus annuus; Lactuca sativa) 7

(10)

Introduction

Black cottonwood (Populus trichocarpa), Green alga (Ostreococcus)

WKKYGNK Moss (Physcomitrella patens) 5

WRKYGQT Brachypodium distachyon 2

WKKYGPK Brachypodium distachyon 1

WHKYGAK Micromonas sp. RCC299 1

WRKYGHK Green alga (Ostreococcus) 1

WRKYGNK Green alga (Ostreococcus) 2

WKNNGNT Alga fungi (Phycomyces), Rhizopus 12

WTKYDQR Strawberry (Fragaria vesca) 1

WREYDQR Strawberry (Fragaria vesca) 1

Total 1975

All WRKY proteins contain either one or two WRKY domains. Based on the number of WRKY domains and the structural features of the zinc-finger-like motif, WRKY protein family was originally divided into three groups. WRKY proteins with two WRKY domains are group I proteins while those with a single WRKY domain are group II or III. Group II WRKY proteins are further subdivided into five subgroups IIa, IIb, IIc, IId and IIe according to the presence of short conserved structural motifs. Group III differs from I and II in its variant C2HC zinc finger motif

CX7CX23HXC (Eulgem et al., 2000).

Some WRKY proteins exist as chimeric proteins combining NBS-LRR (nucleotide binding site - leucine rich repeat) proteins and WRKY domains (Deslandes et al., 2002; Noutoshi et al., 2005; Rushton et al., 2010). AtWRKY52/RRS1 is such a protein that contains a group III WRKY domain C-terminal to a TIR-NBS-LRR (Toll/interleukin-1 receptor-nucleotide-binding site-leucine-riche repeat) domain and mediates R (resistance)-gene based resistance to the bacterial pathogen

Ralstonia solanacearum (Deslandes et al., 2003). In addition, AtWRKY16/TTR1

and AtWRKY19 are also NBS-LRR-WRKY proteins found in Arabidopsis. Other examples are GmWRKY176 from soybean (Glycine max), OsiWRKY41 (DAA05106) from indica rice (Oryza sativa indica) and ABF81432 from black cottonwood (Populus trichocarpa).

(11)

Introduction

Till now, two reports on crystal structure of WRKY proteins are available (Duan et

al., 2007; Yamasaki et al., 2005). The WRKY domain of AtWRKY4 consists of a

four-stranded β-sheet, with the zinc coordinating Cys/His residues forming a zinc-binding pocket. The WRKYGQK residues correspond to the most N-terminal β-strand, which partly protrude from the protein surface and enable access to the major DNA groove during the interaction with the target DNA. This WRKYGQK-containing β -strand seems to make contact with an approximately 6-bp region, which is in consistent with the length of the consensus W-box (Yamasaki et al., 2005; Yamasaki et al., 2008). The crystal structure of the AtWRKY1 C-terminal WRKY domain is very similar with AtWRKY4 and contains an extral β-strands upstream of the WRKYGQK motif, thereby with DNA-binding residues located at the second and the third β-strands (Duan et al., 2007).

1.1.2 Distribution and evolution of WRKY transcription factors

The first WRKY protein described was SPF1 (Sweet-Potato Factor 1) in sweet potato which was found to bind DNA upstream of genes coding for sporamin and beta-amylase (Ishiguro and Nakamura, 1994). Shortly after the first report in sweet potato, some discovery on WRKY proteins was made in other plant species including wild oat (Avena fatua), Arabidopsis thaliana and parsley (Petroselinum

crispum) (de Pater et al., 1996; Rushton et al., 1996; Rushton et al., 1995). Since

then, knowledge about WRKY transcription factors has substantially accumulated (see review Eulgem and Somssich, 2007; Pandey and Somssich, 2009; Ross et

al., 2007; Rushton et al., 2010). The WRKY gene family has been analyzed in a

number of plant species including barley (Hordeum vulgare), bittersweet nightshade (Solanum dulcamara), chamomile (Matricaria chamomilla), Citrus spp, creosote bush (Larrea tridentata), cucumber (Cucumus sativus), grapevine (Vitus

aestivalis), orchardgrass (Dactylis glomerata), potato (Solanum tuberosum and S.chacoense), rice (Oryza sativa), tobacco (Nicotiana tabacum), tree cotton

(Gossypium arboreum), white weeping broom (Retama raetam), tomato (Solanum

lycopersicum) and soybean (Glycine max) (Ross et al., 2007; Rushton et al., 2010;

(12)

Introduction

The number of WRKY proteins is expanding from one in the unicellular green alga

Chlamydomonas reinhardtii to nearly 200 in soybean (Glycine max) (Rushton et al.,

2010). Therefore, it is believed that WRKY proteins had a lineage-specific expansion in plants during the evolutionary process. Compared to the green alga and moss, flowering plants have the largest WRKY family. Due to their wide presence in plants and their unique expansion in dicot and monocot plants, WRKY proteins were initially considered as plant-specific transcription factors. However, the discovery of proteins with WRKY domains in the protist Giardia lamblia and the slime mold Dictyostelium discoideum challenged this concept and implicated a much earlier origin of WRKY proteins (Pan et al., 2009; Ülker and Somssich, 2004). Up to now, proteins with putative WRKY domains have also been found in other non-plant organisms including the zygomycetes Rhizopus oryzae,

Phycomyces blakesleeanus, Mucor circinelloides and the slime mold Dictyostelium purpureum.

Consecutive WRKY domain gain and loss led to an expansion of the WRKY family, and that a rapid amplification of the WRKY genes appeared to be earlier than the divergence of monocot and dicot plants (Wu et al., 2005). Despite some debate on the evolution of WRKY domains, it is now well accepted that group I WRKYs are the most ancient WRKY proteins evidenced from the unicellular green alga

Chlamydomonas reinhardtii. There is evidence supporting a late evolution of group

II, however, group III was also considered as the last evolved group due to its expansion in monocot plants (Mangelsen et al., 2008; Rushton et al., 2010; Ülker and Somssich., 2004; Zhang and Wang, 2005). It was revealed that some sequence-related homologous WRKY proteins have conserved functions between monocots and dicots (Mangelsen et al., 2008; Proietti et al., 2011).

1.2 Biological functions of WRKY transcription factors

Numerous studies have revealed the significance of WRKY transcription factors in multiple processes including development, hormone signalling and responses to biotic and abiotic stresses (Rushton et al., 2010). A single WRKY transcription factor might mediate transcriptional reprogramming associated with multiple signalling pathways. On the other hand, multiple WRKY proteins might act in a

(13)

Introduction

single physiological process synergistically or antagonistically (Pandey and Somssich, 2009). The interconnected signalling network of WRKY factors possesses multiple inputs and outputs (Rushton et al., 2010)

1.2.1 WRKY in biotic stresses

Plants have evolved two layers of defense mechanisms to antagonize the invading pathogens (Jones and Dangl, 2006). These two interconnected branches are termed as MAMP-triggered immunity (MTI) and effector-triggered immunity (ETI) and they are initiated either relying on the recognition of conserved microbial molecules, so-called microbe-associated molecular patterns (MAMPs) or the recognition of pathogen-derived specific (a)virulence factors (effectors). The local and systemic defense responses activated by MTI or ETI often require the modulation from phytohormones such as jasmonate (JA) and salicylic acid (SA) (Bostock, 2005; Durrant and Dong, 2004; Pandey and Somssich, 2009). These responses upon pathogen invasion require massive transcriptional reprogramming, which was achieved by transcription factors including WRKY proteins (Eulgem, 2005; Naoumkina et al., 2008; Ryu et al., 2006; Wang et al., 2006). Such transcriptional reprogramming associated with plant defense leads to timely and balanced activation/repression of diverse targets in plant immune responses. Thus, WRKY factors are considered as central regulators in plant innate immune system (Eulgem and Somssich, 2007).

Gain- or loss-of-function studies have demonstrated that WRKY proteins are critical regulators of plant immune responses either positively or negatively in a sophisticated defense response network (Deslandes et al., 2002; Journot-Catalino

et al., 2006; Kim et al., 2008; Li et al., 2006; Murray et al., 2007; Shen et al., 2007;

Zheng et al., 2007). The R - gene type protein AtWRKY52 confers strong resistance towards the bacterial pathogen Ralstonia solanacearum (Deslandes et

al., 2002). This R-gene mediated resistance was achieved through its nuclear

interaction with the bacterial effector PopP2 (Deslandes et al., 2003). Interestingly, a single amino acid insertion in the WRKY domain led to conditional activation of defense responses and a loss in the DNA-binding capability (Noutoshi et al., 2005). In addition, AtWRKY52 provides dual resistance against fungal and bacterial

(14)

Introduction

pathogens by synergistically acting with the R protein RPS4. AtWRKY70 was described as a shared component in SA and JA-dependent defense pathways and played crucial role in the cross-talk of SA and JA signalling (Li et al., 2004; Li et al., 2006; Ren et al., 2008). Moreover, AtWRKY70 was indispensable for the transduction of R gene-mediated resistance and receptor-like protein (RLP)-mediate immunity (Knoth et al., 2007; Zhang et al., 2010). It was also suggested to positively modulate systemic acquired resistance (SAR) (Wang et al., 2006). Other positive regulators of resistance in Arabidopsis include AtWRKY3, -4 and -33, which play a role in resistance against necrotrophic pathogens Botrytis cinerea and Alternaria brassicicola (Lai et al., 2008; Zheng et al., 2006). In addition, AtWRKY8 was characterized recently as a positive regulator of basal defense to

B.cinerea but a negative regulator to Pseudomonas syringae (Chen et al., 2010a).

Most information about functions of WRKY transcription factors came from the model dicot plant Arabidopsis, but their importance in pathogen defense was also demonstrated in monocots like rice and barley. In rice, majorities of the OsWRKY genes are responsible to pathogen challenge, abiotic stresses and phytohormone treatment (Ramamoorthy et al., 2008; Ryu et al., 2006). Overexpression studies have demonstrated several WRKYs (OsWRKY3, -13, -31, -45, -53, -71 and -89) to be associated with rice resistance towards Magnaporthe grisea and/or

Xanthomonas oryzae pv. oryzae (Xoo) (Chujo et al., 2007; Liu et al., 2005; Liu et al., 2007; Qiu et al., 2007; Qiu et al., 2008a; Qiu and Yu, 2009; Shimono et al.,

2007; Tao et al., 2009; Wang et al., 2007; Zhang et al., 2008). For instance, overexpression of OsWRKY13 enhances resistance to the bacterial blight Xoo and the rice blast M. oryzae by activating the SA synthesis and suppressing the JA pathway (Qiu et al., 2007; 2008a). Similarly, OsWRKY71 was shown to be inducible by SA and overexpression of OsWRKY71 enhanced the rice resistance to Xoo through the indirect activation of OsPR1b and OsNPR1 (Liu et al., 2007). OsWRKY45 plays pivotal role in BTH-induced resistance to rice blast fungus through the SA pathway (Shimono et al., 2007; Shimono et al., 2011).

OsWRKY45-1 (japonica-derived WRKY45) and OsWRKY45-2 (indica-derived WRKY45) overexpression resulted in enhanced resistance to the rice fungal

pathogen M. oryzae, however, they have opposite effects on the resistance to

(15)

Introduction

WRKY45 (WRKY45-1) enhanced susceptibility to Xoo while overexpression of

indica-derived WRKY45 (WRKY45-2) resulted in resistance to Xoo (Tao et al., 2009). Overexpression of OsWRKY53 protected the transgenic rice plants from the blast fungus M. grisea and stimulated the expression of PR proteins and peroxidase enzymes (Chujo et al., 2007).

Depending on the pathogen, WRKY proteins may have contrasting effects on the resistance to diverse pathogens (e.g. the aforementioned AtWRKY8), which is also known for SA- and JA-mediated defense responses. But a number of WRKY transcription factors act, as far as known till now, only as negative regulators in plant defense responses. In Arabidopsis, the sequence-related AtWRKY18, -40 and -60 function in a partially redundant way in negatively regulating resistance to bacterial pathogen Pseudomonas syringae (Xu et al., 2006). The wrky18wrky40 double mutant exhibited enhanced resistance towards the biotrophic fungus

Golovinomyces orontii but enhanced susceptibility to the necrotrophic fungus Botrytis cinerea (Shen et al., 2007; Xu et al., 2006). This mutant executes

exaggerated expression of some defense related genes upon pathogen attack. WRKY40-complementation of the wrky18wrky40 double mutants was able to partially restore susceptibility (Pandey et al., 2010).Therefore, AtWRKY18/40 are assumed to act in a feedback repression system that controls basal defense. Similarly, the barley orthologs HvWRKY1 and HvWRKY2 were shown to act as negative regulators of MTI (Eckey et al., 2004; Shen et al., 2007). The ETI to barley powdery mildew (Blumeria graminis f.sp. hordei) is dependent on the recognition of the fungal effector AVR10 by the resistance protein MLA (mildew-resistance locus A) resulting in a hypersensitive response (HR) to the biotrophic fungus. Interestingly, activated MLA10 translocates from plant cytoplasm into nucleus and interacts with HvWRKY1 and -2, leading to the derepression of MTI. Hence, HvWRKY1 and -2 function as a linker between MTI and ETI. Other negative regulators in Arabidopsis include AtWRKY7, 11, 17, 23, 25, 27, 38, -48, -53 , -58 and -62 (Grunewald et al., 2008; Journot-Catalino et al., 2006; Kim et

al., 2006; Kim et al., 2008; Mao et al., 2007; Mukhtar et al., 2008; Wang et al.,

2006; Xing et al., 2008). AtWRKY38 and -62 negatively regulate the basal resistance to P. syringae and their expressions were modulated by PKS5, a SNF1-related kinase (Kim et al., 2008; Xie et al., 2010). Likewise, AtWRKY7, -11 and -17

(16)

Introduction

play a negative role in the defense responses to P. syringae (Journot-Catalino et

al., 2006; Kim et al., 2006). Several rice WRKY genes including OsWRKY62 and

OsWRKY76 were described as negative regulators in immune responses (Peng et

al., 2008; 2010; Seo et al., 2011). The XA21 protein confers resistance to most

strains of the bacteria Xoo in rice (Song et al., 1995). Transgenic rice plants overexpressing OsWRKY62 or OsWRKY76 are compromised in basal defense and XA21-mediated resistance to Xoo (Peng et al., 2008; Seo et al., 2011).

Recently, expanding reports from other plant species confirmed the importance of WRKY proteins in the regulation of biotic stress responses (Giacomelli et al., 2010; Guo et al., 2011; Ishihama et al., 2011; Li et al., 2010a; Marchive et al., 2007; Molan and El-Komy, 2010; Mzid et al., 2007; Oh et al., 2008; Ramiro et al., 2010; Ren et al., 2010b; Skibbe et al., 2008; Van Eck et al., 2010). Ovexpression of grapevine VvWRKY1 and VvWRKY2 in tobacco plants reduced susceptibility to various fungi (Li et al., 2010a; Marchive et al., 2007; Mzid et al., 2007). CaWRKY1 from pepper (Capsicum annuum) appears to function negatively in the defense based on results from its overexpression and gene silencing (Oh et al., 2008). Expression profile studies confirmed the significance of WRKY factors for pathogen resistances in sunflower and coffee respectively (Giacomelli et al., 2010; Ramiro et al., 2010). In tobacco, WRKY4 and WRKY8 were recently demonstrated as positive regulators in pathogen defense (Ishihama et al., 2011; Ren et al., 2010b). An elegant set of experiments in the native tobacco Nicotiana attenuate showed that two WRKY genes, NaWRKY3 and NaWRKY6, coordinate responses to herbivory (Skibbe et al., 2008). NaWRKY3 is required for NaWRKY6 elicitation by fatty acid–amino conjugates in Manduca sexta larval oral secretions, and gene silencing made plants highly vulnerable to herbivores. Similarly, silencing of

TaWRKY53 in wheat through virus-induced gene silencing (VIGS) resulted in

susceptible phenotype to aphid infestation (van Eck et al., 2010). In the recently sequenced genomes, such as poplar (Populus spp.), sorghum (Sorghum bicolour), papaya (Carica papaya) and moss (Physcomitrella patens), the presence of a large number of WRKY proteins was observed (Pandey and Somssich, 2009). However, their functions in plant immunity are yet to be characterized.

(17)

Introduction

1.2.2 WRKY in abiotic stresses

Though lagging behind the studies in biotic stresses, unravelling the roles of WRKY transcription factors in abiotic stress responses has recently become an active emerging field. First evidence for the involvement of WRKY protein in abiotic stress responses came from expression profile studies (Jiang and Deyholos, 2006; Qiu et al., 2004; Ramamoorthy et al., 2008; Sanchez-Ballesta et

al., 2003; Seki et al., 2002; Zou et al., 2007). WRKY transcription factors are

differentially regulated by abiotic stresses including drought (Mare et al., 2004; Ramamoorthy et al.,2008; Rizhsky et al., 2002; Seki et al., 2002), cold (Huang and Duman, 2002; Lee et al., 2005; Qiu et al., 2004; Zou et al., 2010), heat (Li et

al., 2010b; Qiu et al., 2004; Wu et al., 2009), salt (Jiang and Deyholos, 2006; Jiang

and Deyholos, 2009; Qiu et al., 2004; Wei et al., 2008), nutrient deficiency (Devaiah et al., 2007; Kasajima et al., 2010;) and UV radiation (Wang et al., 2007). Recent functional analyses have provided direct evidences for their roles in abiotic stress tolerance. For example, overexpression of OsWRKY45 in Arabidopsis resulted in enhanced salt and drought tolerance (Qiu and Yu, 2009). Similarly, overexpression of OsWRKY11 under heat shock inducible HSP101 promoter conferred tolerance to heat and drought (Wu et al., 2009). Further examples illustrate that WRKY factors are crucial in reprogramming plants when they are under drought or dehydration stress. The barley HvWRKY38 (also called

HvWRKY1) was inducible by drought and cold (Mare et al., 2004). Its ectopic

overexpression in turf and forage grass (Paspalumnotatum Flugge) enhanced drought tolerance (Xiong et al., 2010). Overexpression of AtWRKY39 increased thermotolerance whereas mutation of AtWRKY39 caused susceptibility to heat stress (Li et al., 2010b). Moreover, the AtWRKY39-mediated thermotolerance appeared to be co-regulated by SA and JA. The AtWRKY25 was also reported to be involved in the heat stress responses (Li et al., 2009). Recently, the important role of AtWRKY63 in ABA response and drought stress was uncovered (Ren et al., 2010a). The AtWRKY63 mutant abo3 showed enhanced sensitivity to ABA treatment and reduced drought tolerance. A good example to elucidate the signalling pathways for WRKY-regulated abiotic stresses is from the study on the resurrection plant Boea hygrometrica (Wang et al., 2009). Galactinol synthase

(18)

Introduction

is inducible by dehydration and ABA. The BhGolS1 promoter contains four W boxes and chromatin immunoprecipitation (ChIP) revealed its in vivo binding with the dehydration and ABA-inducible BhWRKY1 (Wang et al., 2009). This finding provides a link between a dehydration-inducible WRKY factor and a downstream target gene that plays a vital role in drought tolerance whereas in most cases the native downstream target genes are largely unknown.

1.2.3 WRKY in developmental processes

Compared with the studies of WRKY genes in stress responses, fewer reports are available on their roles in development processes such as trichome development, seed germination and senescence. Several evidence suggest that members of WRKY proteins are involved in trichome development (Guillaumie et al., 2010; Ishida et al. 2007; Johnson et al. 2002; Wang et al., 2010),embryo formation (Alexandrova and Conger 2002; Lagace and Matton 2004), seed germination (Jiang and Yu, 2009; Zou et al., 2008), senescence (Hinderhofer and Zentgraf 2001; Miao et al., 2010; Robatzek and Somssich 2001; Robatzek and Somssich 2002; Zentgraf et al., 2010; Zhou et al., 2011), dormancy (Pnueli et al. 2002) and metabolic pathways (Sun et al. 2003).

In Arabidopsis, AtWRKY44, also known as Transparent Testa Glabra2 (TTG2), plays a role in trichome development and tannin synthesis in the seed (Johnson et

al., 2002). Another study provides evidence that it is controlling lethality in

interploidy crosses of Arabidopsis (Dilkes et al., 2008). Recently, Wang et al., (2010) reported the role of WRKY proteins in controlling the secondary cell wall formation and lignifications in dicot plants Medicago truncatula and Arabidopsis. Mutation of AtWRKY12 or the Medicago WRKY gene Mtstp1 initiated pith secondary wall formation and substantially increased the stem biomass. This discovery of negative regulators of secondary wall formation in pith shed lights on the possibility of significantly increasing the biomass in bioenergy crops. In rice, OsWRKY78 was suggested to be a positive regulator in stem elongation and seed development evidenced from semi-drawf and small kernel phenotype in RNAi and T-DNA insertion lines (Zhang et al., 2011). Other examples of development-related WRKYs include MINISEED3 (AtWRKY10) in seed development, VvWRKY2 in

(19)

Introduction

regulation of lignifications and AtWRKY75 in root development (Devaiah et al., 2007; Guillaumie et al., 2010; Luo et al., 2005).

In seed germination, the GA-inducible α-amylase enzymes play central roles in germination and post-germination. One early study revealed that wild oat WRKY proteins (ABF1 and ABF2) could bind to the W-box of the GA-regulated α-Amy2 promoter (Rushton et al., 1995), thus linking WRKY proteins with seed germination. Through transient expression studies, some activators and repressors of GA signalling in rice aleurone cells were identified (Xie et al., 2006; Xie et al., 2007; Zhang et al., 2004; Zou et al., 2008). For instance, rice OsWRKY51 and

OsWRKY71 were found to encode repressors of α-amylase whereas OsWRKY72

and OsWRKY77 appeared to be activators. However, genetic evidence is required to validate their real involvement in seed germination. First direct evidence from

Arabidopsis indicated that AtWRKY2 acted as a mediator in the ABA-dependent

seed germination and postgermination growth arrest (Jiang and Yu, 2009). Similarly, the sequence-related AtWRKY18, -40 and -60 were recently characterized as negative regulators of ABA-signalling during seed germination and postgermination growth, with AtWRKY40 playing a central role (Shang et al., 2010). In Arabidopsis, ectopic overexpression of OsWRKY72 caused retarded seed germination, enhanced sensitivity to ABA and altered expression of auxin-responsive genes (Song et al., 2010).

Senescence in plants is a controlled process involving the timely activation of metabolic pathways through transcription factors including WRKY proteins. The WRKY factors are reported to be the second largest group of transcription factors of the senescence transcriptome (Guo et al., 2004). One well-studied example is

AtWRKY53, which showed a specific expression at the onset of leaf senescence

(Hinderhofer and Zentgraf, 2001). It appeared to directly interact with the MEKK1, an upstream components in MAPK cascade (Miao et al., 2007). Moreover, epigenetic programming was also implicated in the mechanism whereby

AtWRKY53 regulates senescence (Ay et al., 2009). Recently, degradation of AtWRKY53 by E3 ubiquitin ligase UPL5 was found essential in executing the leaf

senescence at the right time frame (Miao et al., 2010). AtWRKY6 and AtWRKY22 are also involved in senescence (Robatzek and Somssich 2001; Robatzek and Somssich 2002; Zhou et al., 2010).

(20)

Introduction

1.3 Transcriptional regulatory network of WRKY 1.3.1 WRKY signalling transduction

Expression of many WRKY proteins is induced by the aforementioned diverse stresses. But WRKY transcription factors are also thought to be regulated or activated on the protein level. Some shared components have been identified or postulated upstream of WRKY proteins, for instance, the receptors for microbial-derived molecular signatures and the mitogen-activated protein kinases (Andreasson et al., 2005; Asai et al., 2002; Fiil et al., 2009; Hofmann et al., 2008; Kim et al., 2004; Koo et al., 2009; Liu et al., 2004). In Arabidopsis, WRKY22 and WRKY29 transcription factors act downstream of the flagellin receptor FLS2, a leucin-rich-repeat (LRR) receptor kinase (Asai et al., 2002).

AtWRKY33 was shown to form nuclear complexes with the MAP kinase MPK4. MAMP perception led to the dissociation of the protein complex and release of AtWRKY33, which activated the expression of the camalexin synthesis gene PAD3. (Qiu et al., 2008b). WRKY38 and WRKY62 were shown to act downstream of cytosolic NPR1 in the regulation of jasmonate-responsive gene expression (Mao et al., 2007; Xie et al., 2010). In addition, phosphorylation appeared to be a very important step in the activation of WRKY protein. The MAP kinase kinase kinase (MEKK1) was found to bind directly to the AtWRKY53 promoter and meanwhile phosphorylate AtWRKY53 protein to take a shortcut in signalling (Miao

et al., 2007).

Very recently, Arabidopsis WRKY33 was shown to be a direct phosphorylation target of MPK3/MPK6 following the infection of B. cinerea (Mao et al., 2011). In tobacco, overexpression of the MAP kinase SIPK triggers cell death through the phosphorylation of WRKY1 (Menke et al., 2005). Moreover, phosphorylation of the

Nicotiana benthamiana WRKY8 by MAPK has an important role in the defense

response through activation of downstream genes (Ishihama et al., 2010).

Induced WRKY expression is often extremely rapid and transient, and seems not to require do novo synthesis of regulatory factors (Eulgem et al., 1999; Hara et al., 2000; Lippok et al., 2007; Rushton et al., 1996). Therefore, many WRKY genes are generally considered as early and intermediate stress responsive genes. This

(21)

Introduction

fast responsive behaviour indicates a role for the WRKY proteins in regulating subsequently activated downstream response genes, which eventually leads to protective reactions in plants. As suggested by the binding preference of WRKY proteins for W boxes, genes containing these promoter elements are possible targets of WRKY proteins. They represent a number of stress-related genes (i.e,

PR genes) and WRKY genes themselves (Eulgem et al., 2000; Yu et al., 2001). A

large body of evidences have confirmed the direct targeting of these genes by WRKY proteins. For example, AtWRKY40 was demonstrated to have direct in vivo interaction with promoter regions of the regulatory gene EDS1, the AP2-type transcription factor gene RRTF1 and JAZ8, a member of the JA-signaling repressor gene family (Pandey et al., 2010). AtWRKY6 was found to positively influence the senescence- and pathogen defense-associated PR1 promoter activity (Rabatzek and Somssich, 2002). In addition, it specifically activates the promoter of a receptor-like protein kinase SIRK likely through direct W-boxes interactions but represses its own promoter activity (Rabatzek and Somssich, 2002).

1.3.2 Mechanisms of WRKY function

WRKY proteins can function as transcriptional activator or repressor. In Nicotiana

benthamiana, ectopic expression of WRKY8 was found to activate defense-related

genes, such as 3-hydroxy-3-methylglutaryl CoA reductase 2 and NADP-malic enzyme (Ishihama et al., 2010). The tobacco NtWRKY6 acts as an activator in the induction of PR1a gene expression by SA and bacterial elicitor (van Verk, et al., 2008). Heterologous expression of OsWRKY6 in Arabidopsis was shown to activate the expression of defense related genes (Hwang et al., 2011). Some WRKY members may possess both capacities. For example, OsWRKY71 and

OsWRKY77 have been shown to act as activators in ABA signalling but as

repressors in GA signalling (Xie et al., 2005). The similar feature was found for AtWRKY6 and AtWRKY53, which activate other promoters but repress their own promoters (Miao et al., 2008; Robatzek and Somssich, 2002).

An elegant model was proposed for the derepression of MTI in barley-Blumeria

(22)

Introduction

HvWRKY2 act as repressors in the basal defense. Following the recognition of

fungal-derived effector AVR10 by the resistance protein MLA10 in the cytosol, the MLA10 protein was activated and translocated into plant nucleus and physically interacted with HvWRKY1/2 repressor. This was thought to remove the repression from HvWRKY1/2 and activate the promoters of unidentified downstream defense genes. Another example is the activation of PcPR1 by PcWRKY1 in parsley (Turck

et al., 2004; Ülker and Somssich, 2004). It was found that W-box elements are

generally occupied by WRKY factors. Upon recognition of an elicitor (MAMP) by the cognate receptor, the MAPK cascade is activated and translocates a protein kinase into the nucleus where it directly interacts with bound WRKY proteins. Thus, these WRKY factors are replaced and released from the bound W-boxes resulting in the activation of PcPR10.

Other possible mechanisms of WRKY functions include the operation through small RNAs (smRNAs) and histone modifications (Kim et al., 2008; Pandey and Somssich, 2009). Small RNAs, including microRNAs (miRNAs) and small interfering RNAs (siRNAs), have been shown to play fundamental roles in the modulation of gene expression. Some WRKY transcription factors are predicted to be targets of certain miRNAs (Pandey and Somssich, 2009). On the other hand, WRKY proteins might regulate smRNA synthesis. Nevertheless, it is a novel field awaiting further advances. Histone modifications through histone deacetylases are often associated with transcriptional repression by reducing the access of DNA for transcription factors (Zhou et al., 2005). The WRKY genes AtWRKY38, -53, -62 and -70 have been implicated in processes involving histone modifications in the fine-tuning of plant senescence and immunity (Ay et al., 2009; Kim et al., 2008; Liu

et al., 2004). The AtWRKY38 and AtWRKY62 function additively as negative

regulators of basal defense and interact with histone deacetylase 19 (HDA19). HDA19 function positively in the basal defense and can abolish the transactivation activity of AtWRKY38 and AtWRKY62.

Due to the enrichment of W-box elements in the promoter region of WRKY genes, WRKY proteins might physically interact with their own promoter or the promoter of other WRKY genes. This auto-regulation or cross-regulation is a common feature for WRKY action. AtWRKY53 was described to involve in both auto-regulation and cross-auto-regulation (Miao et al., 2008). Likewise, the promoter of

(23)

Introduction

parsley PcWRKY1 was found to be bound with WRKY proteins and activation of

PcWRKY1 promoter was achieved through cross-regulation by other WRKY

factors that activate transcription (Turck et al., 2004).

1.4 Basal defense in barley-Bgh interaction 1.4.1 Barley-Bgh interaction

The obligate fungal biotroph, Blumeria graminis f.sp. hordei (DC) Speer (Bgh), is the causal agent of powdery mildew on barley (Hordeum vulgare. L.). Indicated by the name, this fungus infects only barley but not other cereals. The genetics and physiology of barley-Bgh interaction has been studied since one century ago (Biffen, 1907). Upon landing of a conidial spore on the host surface, the fungus builds the first germ tube for attachment on the leaf surface and water uptake, while a secondary germ tube is built for penetration of the host cuticle and cell wall (Thordal-Christensen et al., 1999). By means of hydrolytical and mechanical power, a small amount of germinated conidia spores might break the cell wall barrier and produce a functional haustorium while others fail to penetrate (Pryce-Jones et al., 1999). After successful penetration of the host cell, the fungus has the ability to reprogram the host cell in the sense that it becomes a nutrient sink and supports fungal proliferation (Schulze-Lefert and Panstruga, 2003). One good example is the green island effect on powdery mildew infected leaves (Schulze-Lefert and Vogel, 2000), though the molecular basis of this redefinition of the infected site as a nutrient sink is not fully understood.

The outcome of a fungal penetration attempt on a compatible host relies on the fungal virulence and the defense state of the attacked cell. Any fungal penetration is only successful when it antagonizes the host defense machinery which is evolved in diverse ways. Early defense prevents penetration and is mainly achieved by the formation of cell wall appositions (CWAs). This mechanical and chemical barrier is constituted of 1,3-glucans (callose), silicon, lignin-like material, and various cell wall proteins. The second line of defense inhibits nutrient uptake of haustoria and it is mainly achieved via hypersensitive response (HR) which is featured by a programmed cell death (PCD) of the attacked and/or the neighboring cells. In addition, HR is associated with accumulation of lignin-like material,

(24)

Introduction

release of phytoalexins and other defense compounds which are toxic to the fungi (Oldach et al., 2001; von Röpenack et al., 1998).

Though the molecular mechanism of HR is largely unknown, it is accepted that an HR reaction is typically triggered by effector-activated resistance (R) proteins (Jones and Dangl, 2006). The number of dominant or semi-dominant race-specific R proteins in barley was estimated to be around 85 (Jørgensen, 1994). Recognition of B.graminis f. sp. hordei is mediated by several loci distributed throughout the barley genome which are designated as Ml (Mildew resistance locus) (Jørgensen, 1994). As the most prominent locus, Mla (Mildew resistance locus A) is located on the short arm of chromosome 1H, with approximately 30 alleles that mediate race-specific resistance (Jørgensen, 1994; Wei et al., 1999). Over the last decades, genetic studies in breeding material have identified a large number of functional resistance genes at the Mla locus in breeding material. Distinct from the genetic structure of Mla with multiple alleles at a single locus, putative AVR genes are scattered throughout the B.graminis f. sp. hordei genome, with the cloned AVR10 belonging to a diverse family encoding proteins lacking secretion signals (Ridout et al. 2006; Skamnioti et al. 2008). Alleles of Mla encode cytoplasmic- and membrane-localized coiled-coil (CC), nucleotide binding site (NBS), leucine-rich repeat (LRR) proteins (Halterman and Wise 2004; Seeholzer

et al. 2010; Shen et al. 2003) that translocate into the nucleus after recognition of

a cognate AVR effector from B. graminis f. sp. hordei. Nuclear localization of AVR is required to mediate the hypersensitive response (Shen et al. 2007). This may be dependent on the direct interaction between appropriate MLA and AVR proteins (Seeholzer et al., 2010). Following recognition, the CC domain of MLA interacts with the transcription factors WRKY1 and WRKY2 (WRKY1/2) (Shen et al., 2007).

1.4.2 Germin-like proteins (GLP) in plant immunity

Members of germin-like protein (GLP) genes were originally isolated from germinating seeds and were regarded as specific marker for the onset of germination (Dunwell et al., 2008; Lane et al., 1993; Thompson and Lane, 1980). They belong to the cupin superfamily proteins which exhibit diverse functions (Dunwell and Gane, 1998). GLPs have been identified from a number of plant

(25)

Introduction

species including Arabidopsis (Carter et al., 1998; Membré et al., 1997; Membré et

al., 2000), soybean (Klink et al. 2007), grapevine (Cramer et al., 2007; Ficke et al.,

2002; Godfrey et al., 2007), conifers (Mathieu et al., 2006), Medicago (Doll et al., 2003; Soares et al., 2009) and peanut (Chen et al., 2010b). The majority of germin and GLP studies are focused on Gramineae species such as wheat, barley and maize (Breen and Bellgard, 2010; Dunwell et al., 2008; Lane, 2002).

Apart from the role of GLPs in germination and early development (De Los Reyes and McGrath, 2003; Federico et al., 2006), they are also implicated in abiotic stress responses such as salt, drought and aluminium stress (Cramer et al., 2007; Houde and Diallo 2008; Ke et al., 2009). Accumulating evidence suggests that GLPs are essential players in plant immune system (Breen and Bellgard, 2010; Lane, 2002). Some GLP genes showed induced expression in response to pathogen, herbivores as well as the chemical treatments like salicylic acid, hydrogen peroxide (H2O2), or ethylene (Dumas et al., 1995; Federico et al., 2006;

Godfrey et al., 2007; Lou and Baldwin, 2006; Schweizer et al., 1999; Wei et al., 1998; Zhang et al., 1995; Zhou et al., 1998; Zimmermann et al., 2006).

The direct involvement of GLP in plant defense has been demonstrated in many cases. For instance, overexpression of a wheat germin in sunflower (Helianthus

annuus) enhanced resistance to pathogens (Hu et al., 2003). Silencing of a GLP in

native tobacco Nicotiana attenuata increased the performance of native herbivore (Lou and Baldwin, 2006). In rice, a cluster of GLPs on chromosome 8 was identified to function as the QTL (quantitative trait locus) responsible for broad-spectrum level resistance (Manosalva et al., 2009). In barley, transient overexpression of certain barley GLP subfamilies resulted in enhanced resistance to the powdery mildew fungus, and silencing of GER4 resulted in enhanced susceptibility to the pathogen (Himmelbach et al., 2010; Zimmermann et al., 2006). In both rice and barley, the GER4 subfamily was identified to contribute most to disease resistance. Recently, a germin-like protein was identified as a transcriptional target of the MLA transcriptional regulon based on quantitative time-course expression profile (Moscou et al., 2011). This reflects an overlapping of basal defense process and R gene-mediated signalling.

Germins and GLPs are targeted to cell surface and have oxalate-oxidase (OXOX) activity (Lane et al., 1993; Lane, 2000) or superoxide dismutase (SOD) activity

(26)

Introduction

(Christensen et al., 2004; Godfrey et al., 2007; Zimmermann et al., 2006). Both enzyme activities are linked to the generation of hydrogen peroxide (H2O2), which

has possible roles in a range of defense reactions, including cell wall reinforcement, cell death, and induction of PR gene expression (Alvarez et al., 1998; Bolwell and Wojtaszek, 1997; Chen et al., 1993; Lamb and Dixon, 1997; Olson and Varner, 1993; Thordal-Christensen et al., 1997; Wei et al., 1998).

In barley, eight GLP genes (HvGER4 a-h) are clustered in the GER4 locus and the promoter contains multiple WRKY factor binding sites (W-boxes) (Himmelbach et

al., 2010). Mutational analysis of W-boxes in GER4c promoter-β-glucuronidase

fusions revealed the enhancing effects of W-boxes in the pathogen-induced promoter activity. Enrichment of W-box elements was also observed in the promoter of OsRGLP2 and TaGLP3 (Mahmood et al., 2010), implicating a potential transcriptional regulation of GLP promoters by WRKY proteins.

1.5 Objectives of this study

The main objective of the current study was to identify putative WRKY transcription factors in barley and characterize the functions of the previously identified HvWRKY1 and -2 with a particular focus in the interaction of barley with

Blumeria graminis f.sp. hordei (Bgh). In order to provide an overview of the WRKY

family in barley, whole-genome search was performed to identify putative WRKY transcription factors based on genomic sequence and transcript databases.

The specific aims were:

1). To identify, annotate members of the barley WRKY gene family and analyze their phylogenetic relationship.

2). To analyse the gene structure and function of HvWRKY1 and HvWRKY2.

3). To identify target genes of HvWRKY1 and HvWRKY2 which were suggested to be negative regulators of barley basal defense.

4). To characterize the disease resistance phenotype of HvWRKY2 overexpression barley lines and compare the defense-related gene expression. 5). To identify candidate genes for further studies and genetic approaches which aim at improving broad-spectrum and durable resistance of barley.

(27)

Materials and Methods

2. Materials and Methods

2.1 Plant growth condition and pathogen infection

Blumeria graminis f.sp. hordei (Bgh) race A6 was maintained in a climate cabinet

and propagated on young seedlings of the susceptible barley cultivar ‘Golden Promise’ at 20°C/18°C (day/night) with 60% relative humidity and a photoperiod of 16 h with 240 μmol m2 s-1 photon flux density.

Evaluation of powdery mildew resistance was performed on detached leaves. Plants from the cultivars Golden Promise, Ingrid, BCIngrid mlo-5(I22) and Sultan 5 (Mla12) were grown in spore-free Percival growth chamber under a photoperiod of 16 h with 240 μmol m-2 s-1 photon flux density. Seven days after sowing, the primary leaf was cut and placed in 0.5% water agar medium containing 40 mg/L benzimidazole in a square (10 × 10 cm) petri dish with the adaxial side of the leaf facing upwards. Each petri dish accommodated about 10 leaf segments.

To use freshly produced conidia for inoculation, old conidia spores from the heavily infected Golden Promise seedlings were removed by gentle shaking of the plants 2 days prior to inoculation. A settling tower was used for inoculations. During inoculation, petri dishes containing the leaf segments were placed inside the tower and conidia from Bgh colonized seedlings were blown and allowed to settle for 10 minutes. The density of inoculum was monitored by haemocytometer and was adjusted to 10-15 conidia per mm2_{for macroscopical observation. Five}

days after incubation, the accessions were scored by counting the number of powdery mildew pustules per 2 cm2 of leaf segment, using a magnifying glass (10x). For evaluation of resistance on single cell level with microscopy, the inoculation density was adjusted around 150 conidia spores per mm2. In the promoter studies, a much higher inoculation density (over 200 conidia spores per mm2) was used to activate the HvGER4c promoter.

2.2 Hygromycin-based selection of transgenic plants

Selection of transgenic plants was established and optimized for barley based on the method of Wang and Waterhouse (1997). Leaf segments 2 cm in length from transgenic and non-transgenic barley plants were cut and immediately placed in MS medium containing 200 mg/L of hygromycin, 0.5mg/L 6-BA

(28)

(6-Materials and Methods

media, leaf segments were kept at 24 oC under a long-day photoperiod (16h/8h light dark regime) for one week until bleaching occurred on the wild type plants.

2.3 Isolation of DNA and RNA

DNA was isolated by CTAB method according to a modified protocol of Doyle and Doyle (1987). Briefly, plant material was ground into fine powder in liquid N2 and

transferred into 2.0 ml microcentrifuge tube. 700 μL hot (pre-warmed at 65°C) CTAB Extraction Buffer was added and the tubes were indubated for 25 minutes. 700 μL of Chloroform:Isoamyl-Alcohol (24:1,CIA) was added and mixed by inversion for 5 minutes. The samples were centrifuged at 10000 rpm for 15 minutes under room temperature. Supernatant was transferred to a new Eppendorf tube containing 600μl of CIA, mixed by inversion for 5 minutes and centrifuged at 10000 rpm for 15 minutes (RT). The supernatant was thoroughly mixed with 500 µL of isopropanol and placed on ice for 15 minutes. Supernatant was discarded and pellet was washed with 70% ethanol/10mM NH4OAc. Finally,

dry pellet was resuspended in 100 μL ddH2O. The DNA concentration was

measured by NanoDrop N1000 (peqLab Biotechnologie GmbH, Erlangen). CTAB Extraction Buffer

2% CTAB

20mM EDTA

100mM tris-Cl, pH 8.0 1.4M NaCl

0.2% mercaptoethanol (add prior to use)

Wash Buffer

70% ethanol 10mM NH4OAc

Extraction of total RNA was performed by phenol-chloroform extraction method. Barley leaves from mock treated or powdery mildew infected samples were harvested at the indicated time points and immediately frozen in liquid nitrogen. Leaf samples were crushed into fine powder in liquid nitrogen using mortar and pestle. 1 mL RNA Extraction Buffer was added to the sample and vortexed vigorously. 200 μL chloroform was added and vortexed again. Thereafter, samples

(29)

Materials and Methods

were vortexed for 15 seconds and centrifuged at 13500 rpm for 15 min at 4°C. The supernatant was transferred into a clean Eppendorf tube with 850 uL chloroform and vortexed briefly. Then all the samples were centrifuged at 13500 rpm for 15 min at 4°C. The supernatant was transferred to a new tube with 1 mL 5 M LiCl and mixed by brief vortex. After overnight incubation at -20°C for precipitation of RNA, pellet was spinned down at 13500 rpm for 20 min at 4°C. The supernatant was carefully discarded and the pellet was washed with 70% ethanol by short vortex followed with centrifugation at 13500 rpm for 10 min at 4°C. The washing step was repeated once and the liquid was removed completely. Then, the pellet was air dried under clean bench for 15 min and dissolved in 50 μL H2ODEPC. RNA

concentration was measured by Nanodrop ND-1000 Spectrophotometer (peqLab Biotechnologie GmbH, Erlangen) and the RNA integrity was examined on denaturing 1.5% agarose-gel containing 5% formaldehyde. Trace DNA was removed using 1 μL DNaseI per μg sample RNA prior to cDNA synthesis.

RNA Extraction Buffer 38% phenol 0.8 M guanidin thiocyanat 0.4M ammonium thiocyanat 0.1M sodium acetate, pH 5 5% glycerol 2.4 Expression analysis

For gene expression analysis, Golden Promise and pUbi::WRKY2 plants were inoculated with Blumeria graminis f.sp hordei A6 or mock treated and harvested at 0, 4 and 12 hours. Total RNA was extracted as described in section 2.3. One μg of RNA was reverse-transcribed using Fermentas reverse transcriptase kit (Fermentas, Sankt Leon-Rot) according to the manufacturer’s instruction. The cDNA was diluted 5-fold (estimated equivalent concentration 10 ng/μL) and used for expression analysis with semi-quantitative PCR and quantitative real-time PCR. In the quantitative real-time PCR, the expression level of IGS, synaptotagmin,

HvPR2 and HvPR5 was determined using the 2-∆Ct method (Schmittgen and Livak,

2008). Amplifications were performed with 20 μl SYBR green JumpStart Taq ReadyMix (Sigma–Aldrich, Munich) with 350 nM oligonucleotides and an Mx3000P

(30)

Materials and Methods

amplification was performed with an initial denaturation step at 95°C for 8 min followed with 40 cycles (94°C for 30 s, 60°C for 30 s, 72°C for 30 s, and 82°C for 15 s). Three fluorescent reading were monitored once at 72°C and twice at 82°C during each cycle. Melting curves were determined at the end of cycling to ensure specific amplification. Ct (cycles of threshold) values were determined and processed with the Mx3000P V2 software. For comparison of expression level, Ct values were generated by deducting the raw Ct values of the candidate genes from the respective raw Ct values of the reference gene barley ubiquitin (Accession Nr., M60175).

In the semi-quantitative RT-PCR, the amplification was performed in 25 μL reaction with 20 ng cDNA as template. The barley ubiquitin gene was used as internal control for equal cDNA usage in PCR reactions. Amplification was achieved by incubation in a DNA thermal cycler for 28-32 cycles, each consisting of 30 s of denaturation at 94 °C, 30 s of annealing at 60 °C, and 30 s of extension at 72 °C. Semi-quantitative RT-PCR (25 μL) 2 μL cDNA (10ng/ul) 2.5 μL 10×PCR Buffer 2.5 μL 2 mM dNTPs-Mix 1.5 μL 25 mM MgCl2 1.0 μL forward primer (10 pM) 1.0 μL reverse primer (10 pM) 0.15 μL Taq polymerase (5 U/μl ) 14.35 μL MilliQ-H2O

2.5 Molecular cloning and plasmids constructions

Primer design was mainly performed with the online tool Primer 3 (http://frodo.wi.mit.edu/primer3/). Restriction sites were introduced on 5’ ends of the primers to facilitate cloning when necessary. In this case, 2-4 bp extra protection nucleotides were added at the ends to improve digestion efficiency of PCR products. All primers used in this study were ordered from Eurofins MWG Operon and listed in Appendix 3. The freeware pDRAW32 (http://www.acaclone.com/) was used for vector information management and in

(31)

Materials and Methods

enzymes from Fermentas (Fermentas, Sankt Leon-Rot). Selection of suitable reaction buffer system for double digestion was assisted by the online tool DoubleDigest™ (http://www.fermentas.com/en/tools/doubledigest).

Standard transformation procedure was followed using 60 seconds heat shock of

E.coli strain DH5at 42 oC. Positive colonies grown on antibiotic selection plates

were verified by colony PCR. The colonies were further confirmed by restriction digestion and sequencing (LGC Genomics, Berlin).

Vector constructs were generated following standard cloning procedure. Maps for all the vectors generated in this study were included in Appendix 4. The wild-type and truncated HvGER4c promoter GUS fusion constructs were provided by Dr. Patrick Schweizer (IPK, Gartesleben). Plasmid pGY1-mCherry (4213 bp) was generated from p123mCherry and pGY1-GFP. In principle, GFP in pGY1-GFP was replaced by mCherry released from p123mCherry after NcoI/EagI digestion. For cloning of pUbi::AtWRKY40, AtWRKY40 CDS was amplified by AtW40Bam_F and AtW40Hind_R from pDONR-AtWRKY40, which was kindly provided by Dr. Imre E. Somssich (Max Planck Institute for Plant Breeding Research, Köln). As the stop codon was absent in the original cDNA clone, the codon TAA was added in the reverse primer AtW40Hind_R. Ligation of BamHI/HindIII digested AtWRKY40 PCR product and the backbone pUbi-AB yielded pUbi::AtWRKY40.

HvWRKY1 promoter was amplified using primer pHvW1Bam_F1/ pHvW1EcoR_R

(954 bp) and pHvW1Bam_F2/ pHvW1EcoR_R (1940 bp) from barley (cv. Golden Promise) genomic DNA. PCR products were digested with BamHI/EcoRI and ligated with linearized pGusi-AM (5.3 kb). HvWRKY2 promoter (2876 bp) was amplified by pHvW2Bam_F and pHvW2Hind_R.

To clone artificial microRNA for HvWRKY2 silencing, a 21-bp-long sequence (TTCAGACGTAGTCACCGACTA) was selected by WMD (Web MicroRNA designer, http://wmd.weigelworld.org/cgi-bin/mirnatools.pl) for specific targeting of

HvWRKY2. Four primers including 394ImiR-s, 394IImiR-a,

HvW2-394IIImiR*s and HvW2-394IVmiR*a were used to run PCR using pNW55-osaMIR528 as template (kindly provided by Prof. Detlef Weigel). Three PCR products were fused by the primer pair amiRPCR4_F and amiRPCR4_R. The final PCR product was digested with EagI/SpeI and ligated at the compatible ends of

(32)

Materials and Methods

pGY1-GFP produced by EagI/XbaI digestion. The resulting plasmid was named pGY1-GFP-amiRWRKY2.

2.6 Isolation of plasmid DNA

The recombinant bacterial were cultured with LB-medium including 100 mg/L ampicillin. The mini-preparation of plasmid DNA was performed using PureYield™ Plasmid Miniprep System (Promega) from 4 mL overnight bacterial culture following the manufacturer’s instructions (Technical Bulletin #TB374). For midi-preparation, PureYield™ Plasmid Maxiprep System (Promega) was used for plasmid isolation from 75 mL overnight culture. As the last step, all plasmids were eluted in ddH2O instead of TE buffer to facilitate further analysis. Plasmids

concentration and purity were examined with NanoDrop N1000 spectrophotometer (NanoDrop Technologies, Wilmington, USA).

2.7 Particle bombardment

A home-made helium biolistic system was used for particle bombardment-mediated transient transformation of barley epidermal cell as described previously (Schultheiss et al. 2002; Schweizer et al., 2000). For each shot, 312 µg of 1.1-µm tungsten particles were coated with pUbi-AtWRKY40 (1 µg) together with

35S::GFP (0.5 µg) as a transformation control. In the control bombardment, the

empty vector pUbi-AB (1 ug) was used together with the GFP construct. 24 h after the bombardment, leaf segments were inoculated with Bgh race A6. Inoculation density was adjusted ca. 150 conidia mm-2. The interaction outcome (penetration efficiency, PE) was analysed 48 h after inoculation by fluorescence microscopy. Transformed GFP expressing cells and the presence of haustorium were identified under blue light excitation. Surface structure of the powdery mildew fungus was detected using fluorescence staining with 0.3% calcofluor (w/v in water) for 30 s. Transformed GFP cells were categorized into three groups as penetrated cells that contained a haustorium, cells that were attacked by a Bgh appressorium but did not generate a haustorium, and cells that were not infected by fungus. Cells with more than one haustorium or that contained haustoria but less than fungi attacked were recorded as only one penetrated cell. The penetration efficiency (PE), referring to the haustorium index (%) in the transformed GFP cells was obtained based on a set of a minimum of three experiments each consisting of at least 100

(33)

Materials and Methods

interaction sites. PE was calculated for each experiment as the number of penetrated cells (presence of a functional haustorium) divided by the total number of attacked cells multiplied by 100.

In the promoter study, particle bombardment was performed using the same biolistic system. For normalization of GUS expression, a plasmid containing the GFP reporter under the control of maize ubiquitin promoter was co-bombarded (pUbi-GFP). In the bombardment, a mass ratio of 2:1:1 for pHvGER4c::GUS (or its truncated forms), 35S::WRKY (HvWRKY1 or HvWRKY2) and pUbi::GFP and 2 ug total DNA was adopted. In the control bombardment, same amount of the empty vector pGY1 was used instead of the WRKY constructs. Bombarded leaves were transferred to 0.5% agar plates supplemented with 40 mg/L Benzimidazole and incubated at 18°C for 48 h before adequate inoculation with B. graminis spores.

2.8 GUS assay

Bombarded leaf segments were inoculated with powdery mildew (B.graminis f.sp

hordei A6) 48 h after transformation. The number of GFP expressing cells was first

counted under fluorescence microscope (Zeiss Axioplan Imaging 2) for each shot. Thereafter, the leaf segments were stained histochemically for GUS expression. Leaf segments were placed in 2 mL Eppendorf tubes and GUS staining solution was added until the leaf tissue was immersed. Short vacuum infiltration was performed till the leaves were completely water-soaked. After 24 h incubation in GUS staining solution at 37oC in dark, the solution was removed and GUS-stained leaves were cleared in clearance solution with shaking. The clearance solution was changed once after 48 hours of incubation. Subsequently, the number of GUS cells per bombardment was counted under macroscopy. The obtained numbers of GUS-stained cells were normalized together to the number of GFP expressing cells from cobombarded pUbi-GFP. Eventually, the normalized number of GUS cells per bombardment was taken as a measure for the HvGER4c promoter activity as previously described (Himmelbach et al., 2010). Average values were based on raw data from at least three independent bombardment experiments. GUS staining solution

0.1 M Na2HPO4/NaH2PO4, pH 7.0

(34)

Materials and Methods

0.5 mg ml–1 5-bromo-4-chloro-3-indoxyl-b-D-glucuronic acid,

cyclohexylammonium salt (X-Gluc; Duchefa, Netherlands) 0.1% (v/v) triton X-100

Clerance solution (1L)

250 ml chloroform

750ml technical ethanol

1.5g trichloroacetic acid (TCA)

2.9 Purification of recombinant protein

To produce recombinant HvWRKY2 protein, the full-length coding sequence of

HvWRKY2 was PCR amplified using HvWRKY2Sal_F and HvWRKY2Hind_R. The

fragment was fused to the C-terminal of thioredoxin-6xHis-S-tag in the expression vector pET32a(+) (Novagen) and resulted in pET32a-HvWRKY2. Subsequently, the construct was electro-transformed into E. coli strain BL21 (DE3) pLysS (Stratagene, La Jolla, USA) using Bio-Rad E.coli Pulser Apparatus at 2.5 kV with 0.2 cm cuvettes.

The bacterial clones containing 6xHis-HvWRKY2 were first verified for the rate of protein production and the solubility of the protein using a small scale protein induction. Large scale (1L) protein production was performed in Luria-Bertani (LB) medium overnight under shaking at 37°C. After inoculation of fresh medium with the overnight culture, bacteria were allowed to grow until mid log phase (OD600 of 0.5-0.8) before isopropyl-β-D-thiogalactopyranoside (IPTG) was added to a final concentration of 1 mM and further incubated for 4 hours for protein induction. The bacteria pellets were harvested by centrifugation at 4000 rpm for 20 min at 4°C and dissolved in 30 ml lysis buffer for 30 minutes. The bacterial cells were disrupted by sonication eight cycles with 20 seconds intervals. As the recombinant is present in the inclusion body, the sonicated lysate was centrifuged at 9500 rpm for 30 min at 4°C and the pellet was dissolved in Buffer B and incubated under shaking for 1 hour at room temperature. Afterwards, the cell debris was removed from the lysate solution by centrifugation for 30 min (12000 rpm). The supernatant was collected and stored at 4°C. To prepare the column for purifying the 6× His-tagged fusion protein, 1 ml of Ni-NTA resin (Qiagen, Hilden, Germany) was pipetted into the column that was clamped onto a stand. The resin was allowed to

(35)

Materials and Methods

settle down and once settled the valve was opened to drain off the residual liquids in the column. Thereafter, the supernatant containing soluble protein was loaded onto a Ni-NTA prepared column and washed three times with 4 ml washing buffer and thereafter, the column was eluted three times with elution buffer. Finally, proteins were desalted and concentrated using an ultra-filtrate column (VIVASPIN 6 ml concentrator) with a molecular weight cut-off (MWCO) at 10 kDa (Vivascience, Lincoln, UK) and stored at -80°C. Protein concentration was estimated by Bradford assay. Different concentrations of bovine serum albumin (BSA) were prepared and used to create a standard curve. Purity and integrity of HvWRKY2 recombinant protein was determined by separating protein aliquots using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). After electrophoresis, gel was fixed by fixation solution (one part Acetic Acid, 3 parts Isopropanol and 6 parts Water) for 30 min. Eventually, the gel was visualized with colloidal Coomassie blue (Roth, Karlsruhe, Germany). Staining solution was added to the gel and shaking at room temperature overnight. To minimize the background staining, destaining was performed using destaining solution for 30 minutes.

LB (Lauria Bertani) Broth 1 % tryptone peptone 0.5 % yeast extract 0.5 % NaCl 1% agar lysis buffer (pH 8.0) 50 mM NaH2PO4 300 mM sodium chloride 10% glycine, 1 mg/ml lysozyme, 0.5 mM PMSF Buffer B 10 mM Tris-HCl (pH 8.0) 8 M urea 100 mM NaH2PO4

Annotation, phylogenetic analysis and functional characterization of barley [Hordeum vulgare L] WRKY transcription factors in the interaction with powdery mildew fungus Blumeria graminis